I have to disagree.
If Microsoft was serious about Windows IoT it should have to design it like Windows CE Platform Builder. A constructor of components with some components open sourced so the system can be built as Lego and ported to a new HW platform.
There is no way to protect investments in Windows IoT. You can’t plan for the future as Microsoft can ditch the board. You can’t plan to extend or change HW design as it depends on Microsoft to make the custom build.
If you run into a bug in Windows IoT the whole project depends on Microsoft to fix it and this is not given, some bugs can exist for years which will jeopardize the project or put a drag on it.
Windows IoT works on a small number of prefabricated development boards, some of them with proprietary HW design so it can’t be extended or replicated. There is no way to run Windows IoT on a custom ARM SoC.
This makes Windows IoT useless for commercial SW/HW development. Uncertainty makes it too dangerous. Maybe it can be a replacement for Windows Embedded, nothing more with the current design choice.
I have two customers who have products based on WinCE that are smart
controllers for a device, with software on a PC integrating with the device.
They have both evaluated Windows IoT and have rejected it, this includes
contacting Microsoft and trying to get answers to their questions. I am
not sure how much of this is that they have custom boards. I know they
both felt like they were getting an Azure Cloud Marketing pitch every time
they spoke to Microsoft about IoT, and the cloud is something they are not
ever going to use. In general, their feeling was IoT did not have what
they need, and that Microsoft wasn’t willing to listen to their concerns.
Now neither customer produces a huge number WinCE based boards, but I wonder
how many of these vendors are out there. In both cases they have also told
their customers to avoid Windows 10, because my customers software which has
worked fine for a long time (i.e. at least Vista era if not Windows 2000)
now works differently about every third update to Win10. Since they are in
markets where reliability and predictably are critical, they are looking at
their current products being useless soon. Before someone brings up Windows
10 Enterprise Long Term Servicing Branch they don’t normally sell the
computers their software and boards are installed on and even if they did,
they would be under the minimum, and still facing major testing cycles
faster than they or their customers can afford.
I hope I am wrong, but the strength of Windows for years was the ability for
small market players to use the environment. While no single player would
even appear on Microsoft’s radar, the cumulative effect of all of them used
to be a major contributor to the cash flow to Redmond. I wonder what will
happen when they lose that market.
Don Burn
Windows Driver Consulting
Website: http://www.windrvr.com
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of xxxxx@hotmail.com
Sent: Friday, March 10, 2017 10:55 AM
To: Windows System Software Devs Interest List
Subject: RE:[ntdev] How to get an EV cert for freelancers / small-time
hobbyist developers?
I have to disagree.
If Microsoft was serious about Windows IoT it should have to design it like
Windows CE Platform Builder. A constructor of components with some
components open sourced so the system can be built as Lego and ported to a
new HW platform.
There is no way to protect investments in Windows IoT. You can’t plan for
the future as Microsoft can ditch the board. You can’t plan to extend or
change HW design as it depends on Microsoft to make the custom build.
If you run into a bug in Windows IoT the whole project depends on Microsoft
to fix it and this is not given, some bugs can exist for years which will
jeopardize the project or put a drag on it.
Windows IoT works on a small number of prefabricated development boards,
some of them with proprietary HW design so it can’t be extended or
replicated. There is no way to run Windows IoT on a custom ARM SoC.
This makes Windows IoT useless for commercial SW/HW development. Uncertainty
makes it too dangerous. Maybe it can be a replacement for Windows Embedded,
nothing more with the current design choice.
—
NTDEV is sponsored by OSR
Visit the list online at:
http:
MONTHLY seminars on crash dump analysis, WDF, Windows internals and software
drivers!
Details at http:
To unsubscribe, visit the List Server section of OSR Online at
http:</http:></http:></http:>
All great ntdev threads are digressions.
Why would I use win iot on an rpi product when there are many free and
supported alternatives for rpi that are much lighter weight? I have a win
iot rpi, and indeed it doesn’t actually suck, but I cannot imagine choosing
that as the software to deploy on a real product. It was fun to play around
with, but I have a collection of other small devices none of them running
windows that are at least equally as much fun for kicking bits on.
Mark Roddy
On Fri, Mar 10, 2017 at 10:02 AM, wrote:
>
>
> Talk about thread drift… But, with respect, you’re mixing your concepts
> and as a result talking nonsense.
>
> As a SKU, Windows IoT is definitely not a joke. It’s the replacement for
> Windows Embedded, which has pretty good traction in the market. So, let’s
> start with that. “Windows IoT Enterprise” (for example) is just the new
> name for Windows Embedded Standard, or whatever.
>
> The Galileo? Yes, agreed. Not a useful platform. Sort of a solution in
> search of a problem. Intel trying to be relevant. Kinda sad, I agree.
>
> In terms of Windows in the actual IoT market: Remember, in addition to
> small Intel-architecture systems, Windows IoT runs on a group of ARM-based
> systems, including RPI and various Qualcomm boards. Windows IoT on the RPI
> doesn’t suck. One could make an argument that the developer environment
> (Visual Studio, C#/VB, debugger, etc) coupled with the support of common
> hardware (from RPI out of the box to the Q-Comm systems and BSPs, to other
> more custom platforms), can result in a reasonably compelling package for
> building an IoT product.
>
> Just because, in the past, people have had to cobble together some Linux
> distro to work on some random ARM processor, doesn’t mean it has to be that
> way in the future.
>
> So, no. Definitely not a joke. I’d like to see MSFT increase investment
> in this area. The Microsoft toolchain and development environment has
> turned out to be a major competitive advantage. Who would have thought it
> would happen?
>
> Peter
> OSR
> @OSRDrivers
>
>
> —
> NTDEV is sponsored by OSR
>
> Visit the list online at: http:> showlists.cfm?list=ntdev>
>
> MONTHLY seminars on crash dump analysis, WDF, Windows internals and
> software drivers!
> Details at http:
>
> To unsubscribe, visit the List Server section of OSR Online at <
> http://www.osronline.com/page.cfm?name=ListServer>
></http:></http:>
xxxxx@hotmail.com wrote:
Windows IoT is a joke. I can’t imagine anybody would invest in developing on closed source platform which they unable to tune for a custom platform with a risk of support being discontinued, like happened with Intel Galileo.
The characterization as a “joke” is unfair. I, too, have a little
trouble understanding the decision-making path that would lead to
choosing Windows in my refrigerator or my heart monitor, but the Windows
team has invested vast resources into making that both possible and
practical.
–
Tim Roberts, xxxxx@probo.com
Providenza & Boekelheide, Inc.
>
> This will kill my free software!
Yes. It seems that Microsoft no longer wants the Windows kernel to be
thought of as a hobbyist playground.
–
Tim Roberts, xxxxx@probo.com
Providenza & Boekelheide, Inc.
My current 3 years old SHA1 code certificate expires on August 7st . This will become the date that I will stop Windows developing
software , just like that. The effort and price with the introduction of EV’s and portal submission ( and the time to spend on it )
becomes to high compared to the resulting benefit. I even did not renew my MSDN subsription. I think , some people ( or
companies ) want to get rich too fast.
Christiaan
The message from the powers at be to people like you seems to be “and don’t let the door hit you on the way out”.
You can try use any of vulnerable signed drivers like capcom.sys for loading your own unsigned driver (for instance - https://github.com/Professor-plum/Reflective-Driver-Loader).